flux-controller-patch-releases
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external PRs and commits. \n
- Ingestion points: PR metadata and descriptions are retrieved via
gh pr viewandgit logas described in the Release Flow. \n - Boundary markers: No explicit delimiters are used when interpolating external data into prompts or files. \n
- Capability inventory: The agent can perform sensitive actions including
git push,git tag, andgh pr merge. \n - Sanitization: There is no instruction to sanitize or validate the content of PR titles or descriptions before inclusion in the changelog or commit messages.\n- [COMMAND_EXECUTION]: Utilizes system commands like
gitandghto manage repository lifecycle, including automated branching, committing, and PR management.
Audit Metadata